HyperReCo: Retrieving and Connecting Evidence with Hypergraph Neural Networks for LLM Multi-hop Reasoning
Research
Research News
New papers, breakthroughs and academic AI research
SCLATE: a Substrate for Continual-Learning Agent Training and Evaluation
Research
Supporting and Performing Culture from the Inside
Research
Enabling Timely Guidance before Skill Retrieval: Retaining Helpful Warm Tips in Agent Context
Research
Scoring the Wrong Question: Readout Failures in Constrained-Option Evaluation
Research
From Trajectories to Grounded Preferences: Process Preference Synthesis via Interaction Element Graphs for Web PRMs
Research
LANTERN: Illuminating Hidden Mathematical Knowledge in Language Models
Research
Self-Reconstruction Dynamics for Autoencoder Reconstruction Refinement
Research
Solving Every Step Is Not Enough: Milestone Oracles Reveal a Composition Gap in LLM Math Reasoning
Research
FA-Bench: A Benchmark for Word-Level and Phone-Level Forced-Alignment and ASR Timestamps Under Clean and Noisy Conditions
Research
Before Answering: Evidence Sufficiency under Size-Matched Memory Construction
Research
TRAP: Understanding and Mitigating Privacy Memorization in Language Models
Research